This research will develop a new mathematical model for the analysis of marital fertility. Fertility, a measure of the number of live births per women, will be derived by expanding a model of fecundability (the probability that a woman will conceive per unit time) and will explicitly incorporate age-specific variation in its behavioral and physiological components. The fundamental significance of this research is that any factor that affects female fecundity (and ultimately fertility) must do so through one of the components of this model. Thus the model provides a firm theoretical foundation for a variety of applications in fertility analysis and reproductive biology. Three sources of data will be used to develop and test the model. Contemporary Western populations, for which both detailed demographic and physiological data are available, will provide the basis for development and sensitivity testing. To illustrate the broader applicability of the model, we will also use data from a tribal New Guinea population known to have characteristics very different from those found in Western populations. Finally, the model will be applied to historical data from German villages of the 18th and 19th Centuries. The integrated approach to reproductive research involved in this work brings together expertise in demography, reproductive physiology, and mathematical modeling.